As computer systems grow in speed and shrink in size, power consumed within the computer per unit volume (power density) increases dramatically. Thus, it becomes evermore important to dissipate the heat generated by components within the computer during operation to ensure that the components remain within their normal operating temperature ranges. This reduces a chance that the components fail immediately or have too short a lifetime.
There are two processes by which the heat in computer systems can be dissipated to ensure that the components remain within their normal operating temperature ranges. First, heated air within a chassis of the computer system may be replaced with cool air outside the chassis of the computer system. This is typically known as an air exchange cooling process. Additionally, a specific component may be cooled by directly applying air across the surface of the component. High velocity air immediately applied to the surface of the component raises the convective heat transfer coefficient for the surface of that component, thereby increasing convection cooling with respect to that component. This process is typically known as forced-air-cooling. The computer system may incorporate either process, or a combination thereof, to ensure that the components remain within their normal operating temperature ranges.
In early desktop personal computers, components were passively cooled by radiation or natural convection, the surfaces of the components themselves interfacing directly with still air surrounding the component to transfer heat thereto. Unfortunately, air is not a particularly good conductor of heat. Therefore, in the early desktop computers, the heated air tended to become trapped, creating "hot spots" and acting as a thermal insulator increasing component operating temperature. Eventually, computers were provided with fans to exchange air and force it over the surfaces of the components, decreasing the temperature differential between the surface of the component and the surrounding air to increase the efficiency of heat transfer. The decreased temperature differential overcame some of the poor heat-conducting qualities of air.
Of all components in a computer, the microprocessor central processing unit ("CPU") liberates the most heat during operation of the computer. This springs from its role as the electrical center of attention in the computer. Thus, in prior art computers, motherboards were designed to position the CPU in the flow of air from a cooling fan; other heat-producing components were located away from the CPU to afford maximum cooling of the CPU.
As new generations of microprocessors have arrived, however, this relatively simple scheme has become decidedly inadequate, risking destruction of the CPU. It has become common to associate a heat sink with the CPU to increase the heat-dissipating surface area of the CPU for more effective cooling. Such heat sinks have a plurality of heat-dissipating projections or elements on an upper surface thereof. A lower surface of the heat sink is placed proximate the component and a retention clip is employed to wrap around the heat sink, gripping a lower surface of the component with inward-facing projections.
In addition to the heat sink associated with the CPU, a dedicated CPU cooling fan provides an efficient means to dissipate the heat generated by the CPU. While the primary function of the dedicated fan is to force-cool the CPU, the fan may also act as an air exchanger for the computer system. Typically, such a fan-based system incorporates a "biscuit"-type fan driven by a motor. The dedicated fan in conjunction with the positioning of the CPU within the chassis of the computer system provides the means to force-cool the CPU. Although the fan-based system provides effective component cooling, the fan-based system has draw-backs of its own. Mainly, if the single fan locks up then there is no means to cool the CPU of the computer system because there is no back-up capability built in such fan-based systems. The corollary is that the CPU may overheat causing destruction of the CPU and computer system failure.
A viable solution is to incorporate a secondary, redundant fan into the fan-based cooling system. The evident rational for the redundant fan is to protect the components of the computer system from overheating should the primary fan fail. The redundant fan may be designed strictly as a standby fan, or the redundant fan may be designed to run continuously with the primary fan while the computer system is in standard operation. While the standby fan alterative may provide the greatest back-up security for the fan-based system, this option is not the system of choice for a couple of reasons. First, this option fails to take advantage of the everyday additional cooling capacity the redundant fan offers. Moreover, the market does not support standby systems designed into computer systems without an everyday operational benefit, except in a most critical computer system application.
The second design implementation of a continuously-running redundant fan takes advantage of the additional cooling a second fan offers, while simultaneously fulfilling the ultimate objective for implementing the redundant fan into the fan-based system. Parallel coupling of fans is generally the most efficient in terms of total air flow delivery, but a parallel adaptation is defective for the following reasons.
First, a redundant fan in a parallel position would very likely be by-passed should the primary fan experience a failure. More specifically, the idle primary fan, since it is proximate the secondary fan, short circuits the redundant parallel fan's air flow, because the air takes the path of least resistance. Therefore, the components of the computer system overheat because of the loss of requisite air flow through the computer system. Additionally, a redundant fan in a parallel position does not duplicate the forced-air-cooling capability of the primary fan because the redundant fan is proximate, not axially aligned with, the primary fan. For these reasons, a fan-based cooling system using two fans in parallel does not fulfill the objective for adding a redundant fan and should not be employed in a computer system.
Accordingly, what is needed in the art is a cooling subsystem for a computer system that provides for single fan failure without unduly compromising air exchange or directionality of air flow, thereby maintaining forced-air-cooling of specified components when a single fan has failed. More specifically, what is needed is a cooling subsystem that implements a redundant fan and overcomes the physical limitations inherent in such systems.